Treatment of Elderly Patients with Isolated Systolic Hypertension with Isosorbide Dinitrate in an Asymmetric Dosing Schedule

Home , Isosorbide dinitrate

Journal of Human Hypertension (1998) 12, 557–561  1998 Stockton Press. All rights reserved 0950-9240/98 $12.00 http://www.stockton-press.co.uk/jhh ORIGINAL ARTICLE Treatment of elderly patients with isolated systolic hypertension with isosorbide dinitrate in an asymmetric dosing schedule

MJF Starmans-Kool1,4, HAJ Kleinjans3, FAT Lustermans1, JA Kragten1, JGS Breed2 and LMAB Van Bortel4 1Departments of Internal Medicine and Cardiology, De Wever Hospital Heerlen, 2St Jans Gasthuis Weert, 3Byk Nederland BV, 4Department of Pharmacology, Cardiovascular Research Institute Maastricht University, The Netherlands

Nitrates decrease pulse pressure more than mean (17.9%) than with placebo (5%; P Ͻ 0.05). SBP and MAP arterial pressure (MAP) and are advocated for the treat- decreased compared to baseline, but the changes were ment of isolated systolic hypertension (ISH). Nitrates not statistically significant between the two groups. show drug tolerance during chronic treatment so an DBP tended to increase with ISDN compared to placebo. asymmetric dosing regimen may prevent loss of effect Mean 24-h, mean daytime and mean night-time pulse of nitrates. This study investigates the anti-hypertensive pressure decreased after treatment with ISDN (10.7%, effect of isosorbide dinitrate (ISDN) given in a twice 12.1%, 7.9%, respectively). Pulse pressure tended to daily asymmetric dosing regimen in elderly patients decrease more during the day than during the night with with ISH. ISDN. No changes could be demonstrated with placebo. After a 6-week placebo run-in period, patients entered In conclusion, pulse pressure decreased with ISDN, the double-blind study. Ten patients received placebo resulting in a lower SBP without a decrease in DBP. The and 11 patients ISDN 20 mg b.i.d. for 8 weeks. This dose latter may preserve coronary perfusion in ISH. With the could be doubled once. Office systolic and diastolic asymmetric dosing regimen the decrease in pulse blood pressures (SBP/DBP) and ambulatory BP were pressure was not clear at night. Whether a decrease in measured. Pulse pressure was calculated as SBP–DBP. nocturnal BP, in addition to the spontaneous decrease, Office pulse pressure was more reduced during ISDN is advisable in ISH remains a matter of debate.

Keywords: isolated systolic hypertension; large arteries; anti-hypertensive therapy; nitrates; arterial compliance

Introduction these studies have clearly demonstrated that SBP is an important and independent cardiovascular risk Isolated systolic hypertension (ISH) is characterised factor.2,6,9 by a disproportionate increase in systolic blood In patients with ISH mean arterial pressure (MAP) pressure (SBP) without an elevation of diastolic and vascular resistance may be normal or slightly blood pressure (DBP). In general, ISH is defined by elevated, while pulse pressure, the difference a SBP above 160 mm Hg and a DBP below 90–95 10–12 1 between SBP and DBP, is increased. It has been mm Hg. shown that an increase in pulse pressure contributes Several studies have shown that SBP increases 2,3 considerably to the increase in cardiovascular mor- with age. As DBP decreases after 50 years of age, bidity and mortality seen at higher ages.13 Pulse pulse pressure increases. It has been demonstrated pressure, for a given stroke volume, is determined that the prevalence of ISH increases with age from 3–5 by the ejection rate of the left ventricle, the timing approximately 5% at 60 to 24% at 80 years. Since of reflected waves and the visco-elastic properties of the number of elderly people is still rising, ISH and the arterial wall, such as large artery compliance.12–14 its complications is becoming a major public health In elderly patients with ISH (Ͼ50 years of age) the issue. Epidemiological studies have shown an increase in pulse pressure is mainly due to a direct increased risk of stroke and cardiovascular mor- 2,4,6–8 decrease in large artery compliance and to an bidity and mortality with ISH. In addition, indirect lowering of arterial compliance by an increase in early pulse wave reflections. The SHEP study and SYST-EUR have shown that Correspondence: Dr MJF Starmans-Kool, Department Pharma- 4,8 cology, Maastricht University, PO Box 616, 6200 MD Maastricht, patients with ISH benefit from treatment. Classical The Netherlands anti-hypertensive drugs decrease SBP as well as Received 5 January 1998; revised 22 April 1998; accepted 1 DBP. DBP determines coronary perfusion and thus May 1998 influences an adequate oxygen supply to the heart.15 ISDN treatment of elderly ISH patients MJF Starmans-Kool et al 558 A curvilinear (J-shaped curve) relation has been the last 3 months, unstable angina, severe anaemia, described for the extent of BP reduction through renal or hepatic disease, neurologic abnormalities, anti-hypertensive therapy and the occurrence of pulmonary or endocrine disease except non-insulin myocardial infarction.16,17 The J-curve was seen dependent diabetes mellitus (NIDDM), peripheral especially in elderly patients.18 In ISH patients the arterial disease or treatment with other drugs known coronary system will be frequently narrowed to affect BP. because of atherosclerosis, while left ventricular At randomisation, patients should have a sitting hypertrophy will impair metabolic supply in case of DBP lower than 95 mm Hg and a SBP between 160 a large reduction in DBP.19 Consequently, with the and 200 mm Hg. Twenty-five patients were random- classical anti-hypertensive agents, a further decrease ised, 14 were subsequently treated with ISDN and of DBP in patients with coronary artery disease and 11 with placebo. Laboratory values showed no clini- ISH could be harmful.16,18,20 cally relevant abnormalities at randomisation in all Low-dose nitrates result in a selective dilation of patients. Demographic and haemodynamic para- large arteries, leading to an increase in arterial com- meters at randomisation are shown in Table 1. pliance, while almost no effect on arteriolar tone is present. In addition, nitrates are able to reduce wave Measurements reflection.3,10 As a consequence, nitrates can decrease SBP without a change in DBP.21 Therefore, Office measurements: All haemodynamic measure- nitrates have been advocated for the treatment of ments were performed in a sitting position after at ISH. However, the development of drug tolerance least 10 min of rest in a warm (22 ± 1°C) and quiet may limit the effect and use of nitrates as a therapy room. Measurements were repeated with at least a few for ISH. Using an eccentric dosing regimen, such as minutes interval. The mean of three measurements is used in patients with angina, this tolerance can was used as the average BP value. BP was measured be avoided.22 between 10 and 12 am with a mercury sphygmoman- The aim of the present study was to investigate ometer on the patient’s non-dominant arm. DBP was whether nitrates, given in an eccentric dosing sched- determined at Korotkoff’s phase V. Pulse pressure (PP) ule, can be effective and safe for the treatment of was calculated as SBP–DBP. MAP was calculated as elderly patients with ISH. SBP + 1/3DBP. Heart rate was determined by counting radial pulsations for 30 sec.

Subjects and methods Ambulatory BP monitoring (ABPM): In a subset of Study design centres a 24-h ABPM was performed using a Space- labs 90207 monitor (Spacelabs Inc, Redmond, WA, The study was designed as a multicentre, double- USA). During the day (7 am to 11 pm) BP was blind, placebo-controlled, randomised, parallel recorded every 15 min, during the night (11 pm to study. After a 6-week placebo run-in period, 7 am) every 30 min. Patients were instructed to stop patients were randomised. BP and heart rate muscular activity and to keep their arms entirely measurements were performed at the recruitment quiet during the measurements. visit, 3 weeks before randomisation and at randomisation. Patients were included on the basis of Drug adherence: Patients were asked for their drug BP at randomisation. They were given placebo or adherence at the end of the 8-week treatment period isosorbide dinitrate (ISDN) for 8 weeks. The patients and the number of tablets was counted. started with ISDN 20 mg b.i.d. or placebo. If, after 2 weeks, SBP remained above 160 mm Hg, the dose could be doubled once. To avoid drug tolerance, Data analysis patients were asked to take the medication at 8 am Protocol-correct analysis was performed on all and 2 pm. Before, after 2 weeks, and at the end of patients who completed the study. In addition, an the 8-week treatment period, BP and heart rate intention-to-treat analysis was performed on all ran- measurements were performed. The study was approved by the ethics committees of the different centres and was performed in Table 1 Patient characteristics at randomisation accordance with the Declaration of Helsinki (revised Placebo ISDN version 1983). No. 11 14 Male/female 5/6 5/9 Patients Age (years) 71 (59–80) 69 (54–80) Length (cm) 167 (158–178) 164 (150–188) Thirty-one ISH patients of either sex between 60 and Weight (kg) 67 (53–83) 72 (51–87) 80 years of age were selected from the out-patient SBP (mm Hg) 177 ± 12 184 ± 9 clinics. Demographic data were collected and DBP (mm Hg) 84 ± 585± 5 patients were checked for exclusion criteria. In MAP (mm Hg) 115 ± 5 118 ± 3 ± ± addition, a general physical examination and a rou- PP (mm Hg) 92 4989 Heart rate (beats/minute) 72 ± 677± 12 tine laboratory screening were carried out. Informed consent was obtained from all patients. SBP: systolic blood pressure; DBP: diastolic blood pressure; MAP: Major exclusion criteria were a history of myocar- mean arterial pressure; PP: pulse pressure. Data are mean ± s.d. dial infarction, cardiac surgery or angioplasty within or mean (range). ISDN treatment of elderly ISH patients MJF Starmans-Kool et al 559 Table 2 Ofﬁce measurements

Placebo ISDN

start 2 weeks end start 2 weeks end vs placebo

SBP (mm Hg) 177 ± 13 170 ± 15 166 ± 23* 181 ± 8 167 ± 16 164 ± 16** NS DBP (mm Hg) 84 ± 585± 679± 13 86 ± 486± 486± 7NS MAP (mm Hg) 115 ± 6 113 ± 6 108 ± 15* 118 ± 4 113 ± 7 112 ± 8* NS PP (mm Hg) 92 ± 14 85 ± 17 87 ± 17 96 ± 781± 15 79 ± 15** P Ͻ 0.05 Heart rate 72 ± 672± 667± 7* 77 ± 12 77 ± 11 74 ± 8NS (b/min)

SBP: systolic blood pressure; DBP: diastolic blood pressure; MAP: mean arterial pressure; PP: pulse pressure. Data are mean ± s.d.; *P Ͻ 0.05: difference between 8 weeks of treatment vs start of treatment with ISDN or placebo; **P Ͻ 0.01: difference between 8 weeks of treatment vs start of treatment with ISDN or placebo; NS: not statistically significant. ISDN vs placebo: significant differences between ISDN and placebo after 8-week treatment period. Data after 2 weeks of treatment were not statistically analysed. domised subjects. Data of intention-to-treat analysis in 13 patients. ABPM was recorded in eight patients are shown if different from protocol-correct analysis. of the ISDN group and in five patients of the placebo For comparison of ISDN with placebo after 8 weeks group. SBP was liable to a circadian rhythm with a of treatment statistical analysis was performed using rise in the morning. The highest level was reached the unpaired Student’s t-test. For comparison after around 10 am. When the patients were asleep SBP 8 weeks of treatment with baseline a paired t-test decreased gradually towards its lowest level at was performed. From ABPM data the 24-h mean, approximately 4 am. mean daytime and mean night-time BPs were calcu- Significant reductions in BP were seen in the lated. All data are given as means ± s.d. A P-value ISDN group after 8 weeks of treatment. Mean 24-h Ͻ0.05 is considered statistically significant. values of SBP decreased compared to baseline (7.5%; P Ͻ 0.01) as well as mean daytime (9.6%; P Ͻ Ͻ Results 0.01) and mean night-time SBP (5.1%; P 0.05). For pulse pressure the same pattern was found with Twenty-one patients completed the study, 11 treated a decrease in mean 24-h (10.7%; P Ͻ 0.01), mean with ISDN (nine patients 40 mg b.i.d.; two patients daytime (12.1%; P Ͻ 0.05) and mean night-time 20 mg b.i.d.) and 10 with placebo. Four patients (7.9%; P Ͻ 0.05) values after treatment with ISDN. dropped out of the study after randomisation. Three The decrease in daytime PP tended to be larger than of four were treated with ISDN and stopped because the decrease in night-time PP. DBP did not differ of headache. One patient of the placebo group was statistically from baseline. No statistically signifi- withdrawn because of atrial fibrillation. Within the cant changes in any of the ABPM parameters could group of patients that completed the study, five be demonstrated in the placebo group. experienced an adverse event, four in the ISDN group (two headache, one backpain and one nausea) and one in the placebo group (headache). Discussion Data for the office measurements are shown in In the present study nitrates were used as monother- Table 2. Pulse pressure was significantly more apy for the treatment of ISH. Differences in BP reduced in the ISDN group (17.9%) than in the pla- between ISDN and placebo (vs randomisation) were cebo group (5%; P Ͻ 0.05). Compared to randomis- comparable after 2 and 8 weeks of treatment. There- ation SBP and MAP decreased with ISDN (SBP fore, with the eccentric dosage regimen no tolerance 9.6%; MAP 5.4%) and placebo (SBP 6.1%; MAP effects could be demonstrated within a treatment 6.5%), but the changes were not statistically signifi- period of 8 weeks. Adverse effects were more pro- cant between the two groups. Office DBP tended to nounced in the ISDN group. Most of these patients increase with ISDN compared to placebo. No complained about headache, which is a well-known changes were observed for heart rate. adverse effect with nitrates.23 Data for the ABPM measurements are shown in Apart from the cut-off level for DBP (90 or 95 Table 3 (SBP and PP). This substudy was performed mm Hg), one of the most significant factors affecting

Table 3 ABPM measurements

Placebo ISDN Placebo ISDN ISDN start start end end vs placebo

SBP mean 24 h (mm Hg) 162 ± 24 161 ± 14 154 ± 18 148 ± 13** NS SBP mean day (mm Hg) 165 ± 24 167 ± 14 159 ± 18 151 ± 13** NS SBP mean night (mm Hg) 156 ± 27 149 ± 18 143 ± 20 143 ± 18* NS PP mean 24 h (mm Hg) 81 ± 21 81 ± 13 76 ± 19 73 ± 11** NS PP mean day (mm Hg) 81 ± 21 83 ± 13 78 ± 19 73 ± 10* NS PP mean night (mm Hg) 78 ± 19 78 ± 17 72 ± 20 72 ± 14* NS

SBP: systolic blood pressure; PP: pulse pressure. Data are mean ± s.d. *P Ͻ 0.05: difference between end vs start after ISDN or placebo treatment; **P Ͻ 0.01: difference between end vs start after ISDN or placebo treatment; NS: not statistically significant. ISDN treatment of elderly ISH patients MJF Starmans-Kool et al 560 prevalence of ISH is the number of occasions on inhibitor and a diuretic.4 These data provide evi- which BP is measured.24 With repeated measure- dence that the newer anti-hypertensive agents also ments the prevalence of ISH decreases consider- improve prognosis in ISH. At the moment the ably.6 It is therefore not surprising that in the SHELL trial30 is investigating the effect of long-term present study it was difficult to find patients with treatment with another calcium antagonist persistent ISH and that only a relatively small num- (lacidipine), compared to diuretics, on the incidence ber of subjects could be randomised. of cardiovascular mortality and morbidity over a 5- A method for overcoming transient elevations in year period in ISH patients. BP during clinic measurements is to use 24-h The most important finding of the present study ABPM. In addition, this technique can give infor- was a significant decrease in pulse pressure after mation on circadian changes in the BP profile of ISDN treatment when compared to placebo. Duchier patients with ISH, especially with an asymmetric et al21 have shown that treatment of ISH patients dosing regimen. In the present study several patients with oral sustained-release ISDN also significantly refused a second ambulatory BP recording. There- decreased SBP without a decrease in DBP. These fore, the number of patients was too small to detect effects were sustained after 8 and 12 h of drug any statistically significant difference between the intake. The patients in this study were over 80 years ISDN group and placebo with respect to ambulatory of age, while they were 60–80 years in the current BP data. There was a pattern of a rise in SBP in the study. morning with a decrease at night, comparable to that The decrease in pulse pressure in the present known in normotensive populations. Other studies study is probably achieved by an improvement in have shown that the diurnal BP profile in older arterial compliance31 which, in turn, is mainly patients with ISH is similar to that of other hyper- caused by vasodilation in large arteries. Previously, tensive patients.25 The main determinants of the it has been shown that large arteries (carotid artery, diurnal BP variation in the ISH population were femoral artery, aorta) dilate in response to gender, age, smoking habits and the level of pressure nitrates.31,32 As stated before, the haemodynamic on conventional measurements. Office and ambulat- characteristics of ISH are primarily attributed to a ory systolic and diastolic BPs decreased and tended reduction in arterial compliance rather than to an to decrease, respectively, during the placebo period increase in arterial resistance.28,33 The integration of in the present study. This was not seen in the SYST- cardiovascular reflexes has been found to be dis- EUR study.26 However, in the latter study the pla- turbed in patients with ISH. This also might be due cebo run-in period lasted 3 months. Therefore, in to a decrease in arterial compliance.34 In any case, SYST-EUR a comparison was made between 3 and elevated systolic pressure can be lowered specifi- 4 months of placebo treatment, while in the present cally by an active increase in arterial compliance study the comparison was done between 6 weeks and by reducing pulse wave velocity or reducing and 14 weeks. SYST-EUR did not provide data for wave reflection.12,14,35 Nitrates may actively increase the 3-month run-in period. Consequently, it is diffi- arterial compliance by a reduction in the vascular cult to compare SYST-EUR with the present study. smooth muscle tone of large arteries. In comparison In the present study the decrease in 24-h mean to other vasoactive drugs, nitrates are additionally ambulatory pulse pressure and SBP tended to be capable of a reduction in reflected pressure more pronounced with ISDN as compared to pla- waves10,14,15,36 and therefore could offer further cebo. This is in line with the office measurements. advantages in the treatment of ISH. However, this trend was not present for pulse pres- As well as a decrease in pulse pressure, the sure and systolic pressure at night. The significance present study showed a decrease in SBP, while DBP of an additional decrease in BP at night in combi- tended to increase, thus preserving coronary per- nation with its spontaneous decrease is unknown. fusion. Overall, the haemodynamic effects due to Several studies have shown that all ISH patients treatment with nitrates will decrease cardiac after- are at increased risk for stroke,48 cardiovascular load and may lower the risk for stroke, while the events27 and cardiovascular end-organ damage such risk of decreased coronary perfusion is ruled out. as left ventricular hypertrophy and increased left In conclusion, the present study showed that in ventricular wall stress.19,28,29 Therefore, anti-hyper- elderly patients with ISH pulse pressure decreased tensive therapy has been proposed for patients with with ISDN, resulting in a lower SBP without a persistent ISH. Most experience with treatment of decrease in DBP. This lack of decrease in DBP may ISH involves diuretics.3 The SHEP study (Systolic preserve coronary perfusion in patients with ISH. Hypertension in the Elderly Program) was the first With the asymmetric dosage regimen, the decrease study that demonstrated the efficacy of anti-hyper- in pulse pressure was not evident at night. Whether tensive medication in ISH patients.7,8 In this study a decrease in BP at night, in addition to the spon- treatment with the diuretic chlorthalidone showed a taneous decrease, is advisable in patients with ISH significant reduction in stroke incidence (36%) and remains to be determined. coronary heart disease (27%) over 5 years. Recently, the SYST-EUR study has also shown a significant reduction in stroke (42%) and a decrease in cardio- References vascular morbidity and mortality (31%). These 1 Alderman MH. A review of the Joint National Commit- effects were shown for monotherapy with a calcium tee on detection, evaluation and treatment of high antagonist (nitrendipine) and increased after the blood pressure. The fifth report, 1993. Am J Hypertens addition of an angiotensin-converting enzyme (ACE) 1993; 6: 896–898. ISDN treatment of elderly ISH patients MJF Starmans-Kool et al 561 2 Kannel WB, Dawber TR, McGee DL. Perspectives on of sustained-release isosorbide dintrate for isolated systolic hypertension. The Framingham study. Circu- systolic hypertension in the elderly. Am J Cardiol lation 1980; 61–6: 1179–1182. 1987: 60: 99–102. 3 Staessen J, Amery A, Fagard R. Isolated systolic hyper- 22 Blasini R, Reiniger G, Brugmann U, Rudolph W. Cir- tension in the elderly. J Hypertens 1990; 8: 393–405. cumvention of tolerance development to isosorbide 4 Staessen JA et al. Randomised double-blind compari- dinitrate through use of an interval regimen. Herz son of placebo and active treatment for older patients 1984; 9: 166–170. with isolated systolic hypertension. Lancet 1997; 350: 23 Belz GG, Matthews J, Heinrich J, Wagner G. Controlled 757–764. comparison of the pharmacodynamic effects of nicor- 5 Wing S et al. Isolated systolic hypertension in Evans andil (SG-75) and isosorbide dinitrate in man. Eur J County. I. Prevalence and screening considerations. Pharmacol 1984; 26: 681–685. J Chron Dis 1982; 35: 735–742. 24 Silagy CA, McNeil JJ. Epidemiological aspects of iso- 6 Colandrea MA, Friedman CD, Nichaman MZ, Lynd lated systolic hypertension and implications for future CD. Systolic hypertension in the elderly: en epidemio- research. Am J Cardiol 1992; 69: 213–218. logic assessment. Circulation 1970; 41: 239–245. 25 Thijs L et al. Diurnal blood pressure profile in older 7 Petrovitch H, Vogt TM, Berge KG. Isolated systolic patients with isolated systolic hypertension. J Hum hypertension: lowering the risk of stroke in older Hypertens 1995; 9: 917–924. patients. Geriatrics 1992; 47: 30–38. 26 Staessen JA et al. Ambulatory pressure decreases on 8 SHEP Cooperative Research Group. Prevention of long-term placebo treatment in older patients with iso- stroke by antihypertensive drug treatment in older per- lated systolic hypertension. J Hypertens 1994; 12: sons with isolated systolic hypertension. JAMA 1991; 1035–1039. 265: 3255–3264. 27 Weijenberg MP, Feskens EJM, Kromhout D. Blood 9 Rutan GH et al. Mortality associated with diastolic pressure and isolated systolic hypertension and the hypertension and isolated systolic hypertension risk of coronary heart disease and mortality in elderly among men screened for the Multiple Risk Factor men (the Zutphen elderly study). J Hypertens 1996; 14: Intervention Trial. Circulation 1988; 77: 504–514. 1159–1166. 10 O’Rourke M. Systolic blood pressure: arterial com- 28 Ohtsuka S et al. Alterations in left ventricular wall pliance and early wave reflection, and their modifi- stress and coronary circulation in patients with iso- cation by antihypertensive therapy. J Hum Hypertens lated systolic hypertension. J Hypertens 1996; 14: 1989; 3: 47–52. 1349–1355. 11 Safar ME, Simon AC, Levenson JA. Structural changes 29 Ekpo EB, White AD, Fernando MU, Shah IU. Is iso- of large arteries in sustained essential hypertension. lated systolic hypertension in the elderly more asso- Hypertension 1984; 6(III): III117–III121. ciated with left ventricular hypertrophy and signifi- 12 Simon AC, Levenson JA, Safar ME. Hemodynamic cant carotid artery stenosis than mixed hypertension mechanisms of and therapeutic approach to systolic and isolated diastolic hypertension? J Hum Hypertens hypertension. J Cardiovasc Pharmacol 1985; 7 (Suppl 1995; 9: 809–813. 2): S22–S27. 30 Zanchetti A. Evaluating the benefits of an antihyper- 13 Darne B et al. Pulsatile versus steady component of tensive agent using trials based on event and organ blood pressure: a cross-sectional analysis and a pro- damage: the Systolic Hypertension in the Elderly spective analysis on cardiovascular mortality. Hyper- Long-term Lacidipine (SHELL) trial and the European tension 1989; 13: 392–400. Lacidipine Study on Atherosclerosis (ELSA). J Hyper- 14 Safar ME. Pulse Pressure in essential hypertension: tens 1995; 13 (Suppl 4): S35–S39. clinical and therapeutical implications. J Hypertens 31 Kool MJ et al. Acute and subacute effects of nicorandil 1989; 7: 769–776. and isosorbide dinitrate on vessel wall properties of 15 Safar ME, Levy BI, Laurent S, London GM. Hyperten- large arteries and hemodynamics in healthy volun- sion and the arterial system: clinical and therapeutic teers. Cardiovasc Drugs Ther 1995; 9: 331–337. aspects. J Hypertens 1990; 8 (Suppl 7): S113–S119. 32 Van Bortel LM, Kool MJ, Struijker Boudier HA. Effects 16 Alderman MH, Ooi WL, Madhavan S, Cohen H. Treat- of antihypertensive agents on local arterial distens- ment-induced blood pressure reduction and the risk of ibility and compliance. Hypertension 1995; 26: 531– myocardial infarction. JAMA 1989; 262: 920–924. 534. 17 Cruickshank JM. J curve in antihypertensive therapy. 33 Dart A et al. Aortic distensibility and left ventricular Does it exist? A personal point of view. Cardiovasc structure and function in isolated systolic hyperten- Drugs Ther 1994; 8: 757–760. sion. Eur Heart J 1993; 14: 1465–1470. 18 Coope J, Warrender TS. Lowering blood pressure. Lan- 34 Mancia G et al. Control of circulation by arterial baro- cet 1987; 2: 518. receptors and cardiopulmonary receptors in hyperten- 19 Heesen WF et al. High prevalence of concentric sion. J Cardiovasc Pharmacol 1986; 8 (Suppl 5): remodeling in elderly individuals with isolated sys- S82–S88. tolic hypertension from a population survey. Hyper- 35 Safar M. Treatment of hypertension based on both sys- tension 1997; 29: 539–543. tolic and diastolic pressure could influence the cost of 20 Kannel W, Sorlie P, Castelli WP, McGee DL. Blood therapy. Cardiovasc Drugs Ther 1989; 3: 841–845. pressure and survival after myocardial infarction: the 36 Simon AC et al. Effect of nitroglycerin on peripheral Framingham study. Am J Cardiol 1980; 45: 326–330. large arteries in hypertension. Br J Clin Pharm 1982; 21 Duchier J, Ianascoli F, Safar M. Antihypertensive effect 14: 241–246.